Skip to content
View prasad11s's full-sized avatar

Block or report prasad11s

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
prasad11s/README.md

Hi, I'm Prasad Shimpi

Graduate Research Assistant | MS Applied Data Science @ Syracuse University

I build end-to-end machine learning systems, deploy AI applications, and translate complex data into decisions that matter. My work spans deep learning for space weather forecasting, NLP explainability for financial models, LLM-powered data pipelines, and predictive analytics for agricultural and political systems.

Currently working as a Graduate Research Assistant at Syracuse University's Lender Centre for Social Justice, where I build interactive geospatial dashboards and analyze housing policy data for government stakeholders and community organizations.

Previously worked as a Power BI Developer at Senwell Group, building end-to-end operational data applications using Power Apps, Power Automate, and Snowflake. Before that, as a Data Scientist at ChemView Consulting, I automated consulting report generation reducing manual effort from 3+ hours to under 30 minutes.

I am currently pursuing coursework in Financial Analytics and Advanced Big Data Management.


Featured Projects

Project Description Stack Demo
FinBERT LLM Explainability Comparing SHAP vs LIME on FinBERT for financial sentiment analysis on 9,543 tweets Python, HuggingFace, SHAP, LIME Live Demo
Aurora Geomagnetic Forecasting LSTM neural network for space weather prediction with 48% RMSE improvement over baseline across 1, 3, 6, 12 hour horizons Python, TensorFlow, Keras, NASA OMNI2
Text to SQL with LangChain Query a relational database using plain English via LangChain and Llama 3.3 70B Python, LangChain, Groq, SQLite
Democratic Values & Political Ideology Analysis Cross-national analysis of populism's impact on democratic values across six countries using V-Dem dataset (1970 to 2020) R, rstanarm, Bayesian Analysis, ggplot2
Agricultural Yield Prediction Gradient Boosted Decision Tree model predicting pollution's impact on rice crop yields across Indian districts (13,215 records) R, Shiny, Gradient Boosting, XGBoost
Energy Demand Forecasting Predicting residential energy demand for July using Random Forest and XGBoost in R. R² = 0.478 R, Shiny, Random Forest, XGBoost Live App

Technical Skills

Languages: Python, R, SQL (MySQL)

Machine Learning: Random Forest, Gradient Boosting, XGBoost, Linear/Logistic Regression, Ridge, Lasso, K-Means Clustering, Feature Engineering, Hyperparameter Tuning

Deep Learning: LSTM, CNN, RNN, TensorFlow, Keras, Model Evaluation

NLP & LLMs: FinBERT, LangChain, SHAP, LIME, HuggingFace, Groq API, OpenAI API

Statistical Methods: Hypothesis Testing, Bayesian Analysis (rstanarm, stan_glm), Time Series Analysis, Regression Analysis, ANOVA

BI & Visualization: Power BI (DAX), Tableau, Looker Studio, ArcGIS Online, ArcGIS Experience Builder, ggplot2, Matplotlib, Seaborn

Cloud & Databases: AWS (Cloud Practitioner Certified), Snowflake, MySQL, SQLite

Automation & Development: Power Apps, Power Automate, Streamlit, GitHub

Libraries: Pandas, NumPy, Scikit-learn, XGBoost, Statsmodels


Experience

Graduate Research Assistant | Lender Centre for Social Justice, Syracuse University (Aug 2025 to Present)

  • Built 5+ ArcGIS dashboards translating census and socioeconomic data into accessible visualizations for policymakers
  • Analyzed housing crisis data and drafted policy recommendations for government stakeholder discussions
  • Standardized 10+ census datasets reducing 30+ columns to 10 key indicators per domain

Power BI Developer | Senwell Group Pvt. Ltd (Apr 2024 to Jul 2024)

  • Built end-to-end operational data application using Power Apps, Power Automate, and Power BI with Snowflake integration
  • Developed 8+ interactive dashboards enabling stakeholders to filter and drill down through operational data

Data Scientist | ChemView Consulting Pvt. Ltd (Nov 2023 to Mar 2024)

  • Reduced manual consulting report generation time from 3+ hours to under 30 minutes via Streamlit automation
  • Analyzed 80+ page industry reports across 7 regions to identify automation opportunities

Data Analyst Intern | Data Science Masterminds, Bangalore (Aug 2023 to Nov 2023)

  • Developed performance scorecards and business intelligence dashboards using Power BI, Python, SQL, and Excel
  • Conducted data analysis across multiple client projects, delivering actionable insights through statistical methods and visualization

Currently Working On

  • Graduate Research Assistant at Syracuse University building housing policy dashboards for government stakeholders
  • Coursework: Advanced Big Data Management, Financial Analytics, Lean Six Sigma (Green Belt)
  • Building Streamlit UI for Text to SQL LangChain project with live deployment
  • Developing finance focused projects: stock price prediction, options pricing, portfolio optimization

Education

MS Applied Data Science | Syracuse University (Jan 2025 to Dec 2026) | GPA: 3.61/4.0

Completed: Introduction to Data Science, Database Administration and Management, Quantitative Reasoning, Applied Machine Learning, Deep Learning, Business Analytics

In Progress: Advanced Big Data Management, Financial Analytics, Lean Six Sigma

BSc Data Science | Symbiosis Skills and Professional University, Pune (Aug 2019 to Aug 2022) | GPA: 3.38/4.0


Certifications

  • AWS Certified Cloud Practitioner (2022)

Connect

LinkedIn HuggingFace Gmail


MS Applied Data Science, Syracuse University | Graduating December 2026 | Open to Summer 2026 Internships

Pinned Loading

  1. finbert-llm-explainability finbert-llm-explainability Public

    Comparing SHAP and LIME explainability techniques on FinBERT for financial news sentiment analysis

    Jupyter Notebook

  2. text-to-sql-langchain text-to-sql-langchain Public

    Query a SQLite database using plain English via LangChain and Llama 3.3

    Jupyter Notebook